Optimization of Water Network Retrofit Design for Petroleum Refineries with Water Reuse, Regeneration, and Recycle Strategies (W3R)

Water is a key element in the operation of petroleum refineries. In the past, wastewater
was typically piped to a centralized treatment plant and research efforts were focused
mainly on improving treatment technologies. It was later recognized that distributed
wastewater treatment networks in which wastewater streams are treated separately may
be preferable to the centralized approach. Moreover, scarcities in freshwater supply and
increasingly stringent rules on wastewater discharge have emerges as issues of major
concerns to plant operators, along with an increased awareness in the need to support
sustainable development initiatives and minimization of water footprint. In line with
these development, there are increased interests to incorporate water reuse, regeneration
(i.e, treatment), and recycle (W3R) approaches in the design of refinery water network
systems, with the aim of minimizing freshwater consumption and wastewater
generation. This work presents an optimization model to determine the optimal design
of refinery water network systems. The integrated model explicitly considers the
incorporation of water minimization strategies by first postulating a source-interceptorsink
superstructure that embeds many possible feasible tlowsheet alternatives for the
implementation of potential W3R approaches. Subsequently, a mixed-integer nonlinear
programming (MINLP) model is formulated based on the superstructure to determine
the optimal water network structure in terms of the continuous variables of total stream
tlowrates, contaminant concentrations and the 0-l binary variables of stream
interconnections in the piping network. The superstructure and the MINLP model
explicitly handles the membrane-based interceptors (primarily ultrafiltration and reverse
osmosis units) and the non-membrane-based interceptors, in which in the former, the
feed, permeate, and reject streams are assumed as an individual process units. The
objective of the model is to minimize the fixed capital costs of installing piping
interconnections and the variable cost of operating all stream interconnections while
reducing the pollutants level to within limits by environmental regulations under all the
associated material balances of flows and concentrations. The proposed modeling
approach is implemented on an industrially-significant numerical example using the
GAMS/BARON global optimization platform to obtain a globally cost-optimal water
network topology.